INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

olicy

-0.74

 athlet

-0.69

wana

-0.67

uther

-0.65

 facult

-0.64

asketball

-0.63

ingen

-0.62

azine

-0.61

 Seah

-0.61

POSITIVE LOGITS

ctive

0.67

 retaliate

0.66

pled

0.66

Khe

0.65

FEC

0.64

drawn

0.63

natal

0.60

Zub

0.60

+#

0.60

pling

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.